This proposal seeks renewal of a multi-PI project (Fu and Yeo) to use global approaches to elucidate regulatory principles in the regulation of alternative splicing in mammalian genomes. Built upon our accomplishments in the past funding cycle, including extensive preliminary results, we propose to conduct several systematic loss- and gain-of-function studies to identify genes, gene networks, and pathways involved in the regulation of alternative splicing in three specific aims. The first aim is to perform large-scale network analysis of regulated alternative splicing. Using the two-dimensional mRNA isoform profiling platform developed in our labs, we propose to conduct both genome-wide RNAi and overexpression screening in HEK293 cells and score a set of commonly regulated splicing events (~400) in each treatment condition. We will also complement these genome-wide perturbation studies with transcriptomic analyses by comprehensive RNA-seq against shRNA-mediated depletion (already completed) and ectopic expression of ~300 carefully selected RNA binding proteins (RBPs). These data will help identify new splicing factors, integrate RBPs into transcription, epigenetic and signaling pathways, and decipher both unique and dominant functions of individual RBPs. In the second aim, we propose to analyze RBP-centric protein-protein interaction networks. In particular, we propose to take full advantage of our validated library of open reading frames encoding RBPs for quantitative proteomic analysis of 300 RBPs with and without RNase treatment to identify both RNA-dependent and independent interactions within the framework of RBP complexes. Using the large datasets of functional RNA targets (Aim 1) and RBP-centric protein-protein interactions (Aim 2), we will in Aim 3 perform integrated analysis of RNA genomics data to build predictive models of the regulation of alternative splicing by RBPs. We will use or develop a set of computational tools to predict novel RBP targets, refine individual RBP-central gene networks, and most importantly, integrate comprehensive RBP cis and trans interactomes with system-wide perturbation to build predictive models for cell-specific regulation of alternative splicing. We believe that such integrated analysis will have major impacts on our understanding of regulated splicing and associated disease mechanisms.